检索结果-内蒙古大学图书馆

IEEE Transactions on Audio, Speech and Language processing 2025年 33卷 2045-2058页

作者： Yiming Zhang Xuenan Xu Ruoyi Du Haohe Liu Yuan Dong Zheng-Hua Tan Wenwu Wang Zhanyu Ma Pattern Recognition and Intelligent System Laboratory School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing China Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China Centre for Vision Speech and Signal Processing University of Surrey Guildford U.K. Department of Electronic Systems Aalborg University Aalborg Denmark

In traditional audio captioning methods, a model is usually trained in a fully supervised manner using a human-annotated dataset containing audio-text pairs and then evaluated on the test set from the same dataset. Such methods have two limitations. First, these methods are often data-hungry and require time-consuming and expensive human annotations to obtain audio-text pairs. Second, these models often suffer from performance degradation in cross-domain scenarios, i.e., when the input audio comes from a different domain than the training set, and this issue has received little attention. To address these issues, we propose a new zero-shot method for audio captioning. Our method is built on the contrastive language-audio pre-training (CLAP) model. During training, the model reconstructs the ground-truth caption using the CLAP text encoder. In the inference stage, the model generates text descriptions from the CLAP audio embeddings of given audio inputs. To enhance the ability of the model in transitioning from text-to-text generation to audio-to-text generation, we propose to use the mixed-augmentations-based soft prompt to learn more robust latent representations, leveraging instance replacement and embedding augmentation. Additionally, we introduce the retrieval-based acoustic-aware hard prompt to improve the cross-domain performance of the model by employing the domain-agnostic label information of sound events. Extensive experiments on AudioCaps and Clotho benchmarks show the effectiveness of our proposed method, which outperforms other zero-shot audio captioning approaches for in-domain scenarios and outperforms the compared methods for cross-domain scenarios, underscoring the generalization ability of our method.

关键词： Training Decoding Semantics Data models Acoustics Electronic mail Benchmark testing Transformers Robustness Perturbation methods

来源：评论

学校读者我要写书评

暂无评论

Face recognition with Partial Occlusion Based on Attention Mechanism

Face Recognition with Partial Occlusion Based on Attention M...

引用

2021 International Conference on Electronic information Engineering and Computer Science, EIECS 2021

作者： Zhao, Yuqing Wang, Lin Tan, Mian Yan, Xiaobo Zhang, Xuewen Feng, Hongxi Guizhou Minzu University College of Date Science and Information Engineering Guiyang China Guizhou Minzu University Key Laboratory of Pattern Recognition and Intelligent Systems of Guizhou Guiyang China

ISBN: (纸本)9781665416740

Face recognition is widely used in our life, face recognition with partial occlusion is of great significance to the improvement of face recognition accuracy in practical application scenarios. For improve the accuracy of face recognition in partial occlusion scene, we propose a MobilenetV1-CBAM-Facenet (MC-Facenet) model. which uses Multi-Task Cascaded Convolutional Neural Networks to complete the detection of partial occlusion, then introduces MobilenetV1 and Attention Mechanism to solve the face recognition with partial occlusion. We designs a new Loss function which combines the Triplet Loss and cross-entropy Loss to increase the convergence rate. We carried out comparative experiments based on the public data set LFW, and demonstrated the effectiveness of MC-Facenet under different occlusion types and different occlusion areas. The experimental results show that the occlusion type is glasses and the occlusion area is 30%, the recognition accuracy is improved by 10.92% and 9.39% respectively compared with the original model. It is proved that MC-Facenet can greatly improve the accuracy of local occlusion face recognition. © 2021 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Task-Oriented Multi-user Semantic Communication based on Deep Over-the-Air Computation

引用

IEEE Transactions on Vehicular Technology 2025年

作者： Wei, Shuangying Feng, Chunyan Guo, Caili Zhang, Biling Chen, Jiujiu Beijing University of Posts and Telecommunications Beijing Key Laboratory of Network System Architecture and Convergence School of Information and Communication Engineering Beijing100876 China Nanchang Hangkong University Key Laboratory of Processing and Pattern Recognition of Jiangxi Province School of Information Engineering Nanchang330063 China

The presence of large-scale heterogeneous IoT devices and AI-based applications has brought significant transmission and computation pressure on the wireless communication system. In this paper, we propose a novel multi-user semantic communication framework based on deep over-the-air computation called SC-DAC to tackle this challenge. By integrating communication and computation through a deep over-the-air computation (AirComp) encoder, SC-DAC is able to utilize the superposition of multiple access channels, facilitating multi-user information fusion during transmission and eliminating the need for multi-user detection at the receiver. Further, to achieve the optimal task results, the semantic encoders, decoders, and AirComp encoder are jointly optimized, where stochastic gradient descent (SGD) algorithm and deep double-Q network (DDQN) are employed to optimize the encoder/decoder parameters. The simulation results demonstrate that as the number of users increased, the performance of the system remains relatively stable and shows potential in resource utilization. What's more, when applied to the AI-based tasks such as multi-user multimodal action recognition and semantic segmentation, SC-DAC enhances the accuracy of action recognition by 26% compared with traditional methods and by about 30%-50% compared with other multi-user semantic communication methods under conditions of low signal-to-noise ratio (SNR), while it enhances the accuracy of semantic segmentation by 0.22 to 0.32 compared to other methods © 1967-2012 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Multi-scale Features Fusion Network for Single Image Deraining 10

Multi-scale Features Fusion Network for Single Image Deraini...

引用

10th IEEE Joint International information Technology and Artificial Intelligence Conference, ITAIC 2022

作者： Lai, Yanming Li, Qishen Huang, Hua Li, Qiufeng School of Information Engineering Nanchang Hangkong University Jiangxi Nanchang China School of Software Nanchang Hangkong University Jiangxi Nanchang China Key Laboratory of Jiangxi Province for Image Processing and Pattern Recognition Jiangxi Nanchang China Nanchang Hangkong University Ministry of Education Key Laboratory of Nondestructive Testing Jiangxi China

ISBN: (数字)9781665422079

ISBN: (纸本)9781665422079

Single image rain removal is an important research direction in the field of computer vision. In this paper, the Multi-scale Features Fusion Network (MFFN) is presented for rain removal. MFFN is mainly composed of Multi-features Fusion Module (MFM) and Dual Attention Module (DAM). In the MFM, we make the improved dense block and the dilation convolutions to form the feature extraction branches, which is conducive to improve the receptive fields of network. Subsequently, the bottom-up connection method is adopted between branches to help different branches make full use of image information. Then, feature branches are merged to fusion different features. DAM is consisted of purposed SA Block and standard SE Block. The purpose of DAM is to reduce the non-rain features which is extracted by the network. Experiments show that MFFN can obtain the better result of rain removal than several advanced methods. © 2022 IEEE.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Design of Ultra-Wideband Terahertz Wavelength Division Multiplexer Based on Inverse Design Algorithm

Design of Ultra-Wideband Terahertz Wavelength Division Multi...

引用

2022 International Applied Computational Electromagnetics Society Symposium, ACES-China 2022

作者： Xiao, Zhicheng Song, Kaihong Fang, Ming The Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui University Anhui Hefei230039 China Anhui Province Key Laboratory of Target Recognition and Feature Extraction Anhui Luan237000 China The Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University Hefei230039 China State Key Laboratory of Millimeter Waves Southeast University Nanjing214135 China

ISBN: (纸本)9781665452366

Integrated photonic gadgets play a key function in lots of fields. The process of designing silicon optical devices mostly relies on experience to manually adjust the subtle parameters, which is laborious and ineffective. The proposed inverse design algorithms can simplify the design process of silicon optical devices, improve the design efficiency, and realize multifunctional compact devices. In this paper, two inverse design based methods are proposed to design terahertz band wavelength division multiplexer(WDM). One is based on topology optimization to design a 4-channel ultra-wideband WDM and the other employs a direct binary search technique to create a 3-channel ultra-wideband terahertz WDM, both of which have high transmission, low interference and high integration. Multifunctional photonics devices are predicted to be increasingly used in modern photonic integrated circuits due to these advantages. © 2022 IEEE.

关键词： Optical devices

来源：评论

学校读者我要写书评

暂无评论

Novel Regularization Method for Reduced Biquaternion Neural Network

SSRN

引用

SSRN 2023年

作者： Gai, Shan Huang, Xiang School of Information Science and Engineering Yanshan University Hebei Qinhuangdao066004 China Key Laboratory of Jiangxi Province for Image Processing and Pattern Recognition Nanchang Hangkong University Jiangxi Nanchang330063 China

A reduced biquaternion neural network (RQNN) is a new type of neural network framework that has achieved significant success in machine learning. However, as the reduced biquaternion algebra system contains infinite zero divisors, the RQNN is easily trapped in a local minimum and overfitted. We propose a new regularization scheme for the RQNN to address these issues. First, we propose a new operation in the reduced biquaternion domain—the reduced biquaternion complex modulus (RQCM). Second, we analyse the algebraic properties and obtain the geometric meaning of the RQCM. Finally, we design an improved weight decay method using the RQCM. Unlike conventional weight decay methods with matrix norm constraints, the proposed method can better project the reduced biquaternion in terms of the scale and phase. Colour image classification and denoising tasks are used to demonstrate the superior performance of the RQCM over state-of-the-art weight decay methods. The source code can be found at https://***/tasteimage/Reg4RQNN . © 2023, The Authors. All rights reserved.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

Broadband infrared absorber based on metamaterial

Broadband infrared absorber based on metamaterial

引用

2022 International Applied Computational Electromagnetics Society Symposium, ACES-China 2022

作者： Kang, Meijun Feng, Jian Rong, Mandi Fang, Ming The Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui University Anhui Hefei230039 China Anhui Province Key Laboratory of Target Recognition and Feature Extraction Anhui Luan237000 China The Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University Hefei230039 China State Key Laboratory of Millimeter Waves Southeast University Nanjing214135 China

ISBN: (纸本)9781665452366

In order to obtain an absorber with high absorption efficiency and wide absorption band, a metamaterial structure by Metal-ENZ-Insulator-Metal (MEIM) is designed, and its absorption in the 130-230 THz band is calculated using the simulation software frequency-domain solver. By rotating the upper metal layer and the lower hollow layer, the simulation results can be slightly disturbed. In addition, the introduction of the epsilon-near-zero (ENZ) media will also affect the bandwidth and absorption of the absorber. Therefore, the high absorption and wide bandwidth of the absorber are achieved by the size of the structure, the rotation angle of the corresponding structure,and the clever application of the ITO layer. © 2022 IEEE.

关键词： Metamaterials

来源：评论

学校读者我要写书评

暂无评论

Cd-Vae: An Unsupervised Disentangled Representation Learning Framework for Visual Data

SSRN

引用

SSRN 2023年

作者： Huang, Chengquan Cai, Jianghai Luo, Senyan Wang, Shunxia Yang, Guiyan Lei, Huan Zhou, Lihua Guizhou Key Laboratory of Pattern Recognition and Intelligent System Guizhou Minzu University Guiyang550025 China School of Data Science and Information Engineering Guizhou Minzu University Guiyang550025 China

Disentangled representation learning is a crucial research problem in the field of artificial intelligence, and learning meaningful representation of visual data is useful for improving the generalizability and interpretability of machine learning models. To discover and learn the interpretable factors behind the visual data, previous approaches typically use extra regularization terms in learning disentangled representations of the generated images, which usually lead to poor results between disentanglement and generative quality. For the generative model variational auto-encoder (VAE), the traversal of the generated images along different factor directions shows meaningful and interpretable variations in the latent space. Building upon this observation, we argue that it is important to learn disentangled representations of the generated images by exploiting the scalability and training stability of the generative model VAE, and focusing on meaningful traversal directions in the latent space. Therefore, we propose contrast disentanglement based on VAE (CD-VAE), a deep unsupervised disentangled representation learning framework for visual data, that can discover the latent factors in image data. CD-VAE jointly learns disentangled representations by traversing image variations in the latent space, and constructing contrast disentanglement in the variation space. Experimental results demonstrate that CD-VAE can discover interpretable directions and achieve better disentangled representation learning for visual data. Source codes and pre-training data will be publicly available at GitHub website. © 2023, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Rephrasing the Reference for Non-Autoregressive Machine Translation

arXiv

引用

arXiv 2022年

作者： Shao, Chenze Zhang, Jinchao Zhou, Jie Feng, Yang Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China Pattern Recognition Center WeChat AI Tencent Inc China

Non-autoregressive neural machine translation (NAT) models suffer from the multi-modality problem that there may exist multiple possible translations of a source sentence, so the reference sentence may be inappropriate for the training when the NAT output is closer to other translations. In response to this problem, we introduce a rephraser to provide a better training target for NAT by rephrasing the reference sentence according to the NAT output. As we train NAT based on the rephraser output rather than the reference sentence, the rephraser output should fit well with the NAT output and not deviate too far from the reference, which can be quantified as reward functions and optimized by reinforcement learning. Experiments on major WMT benchmarks and NAT baselines show that our approach consistently improves the translation quality of NAT. Specifically, our best variant achieves comparable performance to the autoregressive Transformer, while being 14.7 times more efficient in inference. Copyright © 2022, The Authors. All rights reserved.

关键词： Neural machine translation

来源：评论

学校读者我要写书评

暂无评论

Investigation on ULA Fitting Promoting Low Coupling Sparse Arrays

Investigation on ULA Fitting Promoting Low Coupling Sparse A...

引用

2022 International Applied Computational Electromagnetics Society Symposium, ACES-China 2022

作者： Yao, Yichang Shi, Wanlu Li, Yingsong Huang, Zhixiang Yang, Guohui Zhang, Kuang Harbin Engineering University College of Information and Communication Engineering Harbin150001 China Anhui Province Key Laboratory of Target Recognition and Feature Extraction LuAn230088 China Anhui University Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education China School of Electronics and Information Engineering Harbin Institute of Technology Harbin150086 China

ISBN: (纸本)9781665452366

A low coupling sparse array (LCSA) is presented, analyzed and discussed on the basis of the recent proposed uniform linear array (ULA) fitting scheme with close-expressions, and its performance with different uniform subarrays in the basic layer is investigated via parameter estimation like direction-of-arrival(DOA). The results gives a superior performance for the presented LCSA. © 2022 IEEE.

关键词： Low mutual coupling sparse array uniform linear array fitting

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：